Andrew Wong

Who they are

Wong's affiliation on the 2021 paper is the Department of Internal Medicine, University of Michigan Medical School, Ann Arbor. The paper was written with eleven co-authors, with Karandeep Singh as the last author.

This page records only what the paper and its published record state. Biography beyond the cited work is not given here.

Key papers

External validation of a widely implemented proprietary sepsis prediction model in hospitalized patients (2021), with Otles, Donnelly, Krumm, McCullough, DeTroyer-Cooley, Pestrue, Phillips, Konye, Penoza, Ghous and Singh. The study examined the Epic Sepsis Model, a proprietary tool that calculates a risk score every 15 minutes. It covered 27,697 adult patients and 38,455 hospital stays at Michigan Medicine between December 2018 and October 2019. Sepsis occurred in 2,552 of those stays. The model's overall ability to tell sepsis cases from other patients was poor, with an area under the curve of 0.63. It did not identify 1,709 of the sepsis cases, 67 per cent, while generating alerts for 6,971 of all 38,455 hospitalised patients, 18 per cent. Compared with ordinary clinical care, it picked out only 183 sepsis patients who had not already received timely antibiotics.

Key ideas

The paper is a model example of external validation: testing a predictive tool on new patients, in a new setting, rather than on the data its developer used. The authors noted that the model's ability to identify sepsis had not been adequately evaluated despite its widespread use, and they raised a second concern alongside missed cases, which is alert fatigue. A warning system that fires for nearly one patient in five gives clinicians many alarms to dismiss, and few of them lead anywhere.

The study is often cited as a case where a clinical prediction tool spread faster than independent testing of it.

---